2D–3D registration remains brittle in challenging scenarios such as low overlap, occlusion, repetitive structures, and severe cross-modal ambiguity. A key reason is that existing methods improve re...
<p><strong>Tác giả:</strong> Chongjian Wang, Junjie Gao</p>
<p>2D–3D registration remains brittle in challenging scenarios such as low overlap, occlusion, repetitive structures, and severe cross-modal ambiguity. A key reason is that existing methods improve representation learning, correspondence estimation, or pose computation in isolation, while the dominant failure mode is inherently cross-level, where errors propagate between features, correspondences, and pose. To address this limitation, we propose DMM-Align: Diffusion-based Matching Matrix Alignme...</p>
<p>📄 DOI: <a href="https://doi.org/https://doi.org/10.1145/3767308.3835975" target="_blank" rel="noopener">https://doi.org/10.1145/3767308.3835975</a></p>
<p>📖 <a href="https://doi.org/https://doi.org/10.1145/3767308.3835975" target="_blank" rel="noopener">Xem bài báo đầy đủ</a></p>
Bài viết được biên soạn lại bởi LIFVietnam. Nguồn tham khảo: OpenAlex
